FIG. 1 shows a phase control thyristor according to a known implementation described in U.S. Pat. No. 5,710,442. In U.S. Pat. No. 5,710,442 a phase control thyristor (PCT) 10 is described with a wafer 2, on which on a cathode side 31 a cathode contact 3 can be arranged. An anode contact 4 is formed on an anode side 41 of the wafer opposite the cathode side 31. Within the wafer 2, there is an (n−) doped drift layer 26 arranged. On this drift layer 26 towards the cathode side 31 is a p doped base layer 5 provided, which contacts the cathode contact 3. (N+) doped cathode layers 7 and (p+) short regions 8 are embedded into the base layer 5. They also contact the cathode electrode 3. Lateral to the cathode contact 3 and separated from it by the drift layer 26 is a gate contact 95 arranged.
On the anode side 41, a p doped anode layer 6 can be arranged, which contacts the anode electrode 4. The cathode and anode sides 31, 41 shall be the plane, at which the outer side of the base layer 5 and the anode layer 6, respectively, are positioned and on which plane the contacts 3, 4 are arranged.
The wafer 2 has (e.g., comprises) an inner region 22 with a wafer thickness 23 measured between the cathode and anode side 31, 41 and a termination region 24, which surrounds the inner region 22. In the termination region 24, there are p doped first and second edge layers 58, 68 arranged with a lower doping concentration than the base layer 5 and the anode layer 6, respectively. The edge layers 58, 68 also terminate in a smaller depth 59, 69 than the base layer 5 and anode layer 6, respectively.
The base layer 5 is arranged up to a base layer depth 51, which depth is measured from the cathode side 31 and in orthogonal projection to it. The base layer 5 and first edge layer 58, which are arranged on the cathode side 31, have a base layer depth 51 and the first edge layer depth 59. The depths 51, 59 are measured as the distance of the maximum depth, to which the layers 5, 58 extend from the cathode side 31. The anode layer depth 61 and the second edge layer depth 69, which second edge layer 68 is arranged on the anode side 41, are measured from the anode side 41.
The edge layer thicknesses 50, 60 are measured as the maximum thickness of the layer in orthogonal projection to the cathode (or anode) side 31, 41. The edge layers 58, 68 have a thickness, which reduces continuously with rising distance from the inner region 22, e.g., the edge layers have a negative bevel 53. The surface of the wafer in the termination region 24 thereby forms a negative angle with the plane of the cathode side 31 or anode side 41, respectively.
As the angle, at which the thickness of the edge layers 50, 60 are reduced, is small (about 2°) in order to reduce the electric field slowly towards the lateral edge of the device, and as the base layer depth 51 is chosen to be deeper than the first edge layer depth 59, high losses are created in the inner region 22 (active area) of the device.
U.S. Pat. No. 7,187,058 describes a power diode with a steep negative bevel. The p+ doped base layer reaches to the edge of the device. A termination layer, which is weakly doped extends to the backside of the device such that it reaches the n+ doped layer at the backside.
Such a device has a “Junction Termination Extension” (JTE). The space charge region of the reverse biased p−n junction is being depleted along the low-doped termination layer towards the backsided n+ layer at the edge, which is the so-called Channel Stop.
The negative bevel makes the termination layer longer compared to a vertical device, which gives a higher breakdown voltage. The necessity to extend the termination layer to the opposite side means that it can only be applied to diodes, because of the diodes having an n doped layer on the backside and it means that the bevel is very steep in order to extend to the backside.
In EP 0 485 059 a known diode having a p+ type anode layer of 14-70 μm is described without disclosing any beveled termination structure. Also US 2007/108558 describes another known diode having an anode layer of 3 μm without beveled termination structure.
In U.S. Pat. No. 4,079,403 a thyristor with a one sided negative bevel which again extends to the other main side of the device, e.g., the device has a steep termination bevel angle. The base layer and anode layer extend to the lateral edges of the device. The layers are terminated in a constant depth from the first main side (which first main side is given as that plane to which the front sided surface of the wafer extends in the inner region).